Computer Science > Machine Learning
[Submitted on 16 Aug 2021 (v1), last revised 12 Jul 2022 (this version, v3)]
Title:On the Opportunities and Risks of Foundation Models
View PDFAbstract:AI is undergoing a paradigm shift with the rise of models (e.g., BERT, DALL-E, GPT-3) that are trained on broad data at scale and are adaptable to a wide range of downstream tasks. We call these models foundation models to underscore their critically central yet incomplete character. This report provides a thorough account of the opportunities and risks of foundation models, ranging from their capabilities (e.g., language, vision, robotics, reasoning, human interaction) and technical principles(e.g., model architectures, training procedures, data, systems, security, evaluation, theory) to their applications (e.g., law, healthcare, education) and societal impact (e.g., inequity, misuse, economic and environmental impact, legal and ethical considerations). Though foundation models are based on standard deep learning and transfer learning, their scale results in new emergent capabilities,and their effectiveness across so many tasks incentivizes homogenization. Homogenization provides powerful leverage but demands caution, as the defects of the foundation model are inherited by all the adapted models downstream. Despite the impending widespread deployment of foundation models, we currently lack a clear understanding of how they work, when they fail, and what they are even capable of due to their emergent properties. To tackle these questions, we believe much of the critical research on foundation models will require deep interdisciplinary collaboration commensurate with their fundamentally sociotechnical nature.
Submission history
From: Rishi Bommasani [view email][v1] Mon, 16 Aug 2021 17:50:08 UTC (16,173 KB)
[v2] Wed, 18 Aug 2021 17:07:22 UTC (16,193 KB)
[v3] Tue, 12 Jul 2022 23:45:14 UTC (40,407 KB)
Current browse context:
cs.LG
References & Citations
DBLP - CS Bibliography
export BibTeX citation
Loading...
Bibliographic and Citation Tools
Bibliographic Explorer (What is the Explorer?)
Connected Papers (What is Connected Papers?)
Litmaps (What is Litmaps?)
scite Smart Citations (What are Smart Citations?)